Geomorphological and ecological characterization of a hard-substrate complex in a temperate shallow shelf sea (SE North Sea) using hydroacoustic sensors and machine learning algorithms

Author(s):  
Rune Michaelis ◽  
Lasse Sander ◽  
Finn Mielck ◽  
Svenja Papenmeier ◽  
H. Christian Hass

<p>The North Sea is a shallow marine environment. The sediment distribution of the seabed is dominated by sand-sized material. Hard-substrate areas are a relatively rare, but important habitat for sessile and mobile species. This habitat type forms island-like geomorphic features owing to the presence of glacial deposits in the shallow subsurface. While their ecological importance is widely acknowledged, hard-substrate areas are characterized by a large degree of spatial heterogeneity and an unaccounted high local diversity in physical surface properties, sediment composition and temporal change.</p><p>The aim of this study is the detailed investigation into the spatial characteristics and temporal variability of an exemplary hard-substrate complex located 10 km offshore the island of Sylt (N-Germany). The area has a size of c. 3 km<sup>2</sup>and was investigated between 2008 and 2019 using a range of hydroacoustic and optical sensors (multibeam echosounder, sidescan sonar, sub-bottom profiler, acoustic ground discrimination system, underwater videos) and machine learning algorithms (haar-like features) to track the changes in the number and local distribution of exposed stones.</p><p>The maximum water depth in the area is 16 m and a linear arrangement of hard substrates emerges up to 4 m from the seabed. A layer of fine sand with a thickness of 0.5 m overlays the more planarly deposited coarse sediments in the proximity of the stony outcrop. This layer of fine sand is relatively mobile and leads to a frequent temporal change of the distribution of sediment on the seabed, whilst the stony outcrop is only marginally affected by the spatial dislocation of sediments. The spatial extent of hard substrates is variable due to the presence of a mobile sand cover on the seabed.</p><p>This study emphasizes the need for quick and automated object classification routines to be integrated in monitoring approaches in the highly dynamic coastal zone. It has shown that the geomorphological diversity and interannual variability of hard-substrate areas can be captured using the presented approach. Detailed studies and monitoring tools are important to better understand the interrelation of geomorphological and sedimentary processes at the seabed with the ecology of epibenthic organisms.</p><p><strong>Keywords: </strong>North Sea; hard-substrate habitats; mobile sediments; hydroacoustic; haar-like features</p>

2020 ◽  
Vol 39 (5) ◽  
pp. 6579-6590
Author(s):  
Sandy Çağlıyor ◽  
Başar Öztayşi ◽  
Selime Sezgin

The motion picture industry is one of the largest industries worldwide and has significant importance in the global economy. Considering the high stakes and high risks in the industry, forecast models and decision support systems are gaining importance. Several attempts have been made to estimate the theatrical performance of a movie before or at the early stages of its release. Nevertheless, these models are mostly used for predicting domestic performances and the industry still struggles to predict box office performances in overseas markets. In this study, the aim is to design a forecast model using different machine learning algorithms to estimate the theatrical success of US movies in Turkey. From various sources, a dataset of 1559 movies is constructed. Firstly, independent variables are grouped as pre-release, distributor type, and international distribution based on their characteristic. The number of attendances is discretized into three classes. Four popular machine learning algorithms, artificial neural networks, decision tree regression and gradient boosting tree and random forest are employed, and the impact of each group is observed by compared by the performance models. Then the number of target classes is increased into five and eight and results are compared with the previously developed models in the literature.


2020 ◽  
pp. 1-11
Author(s):  
Jie Liu ◽  
Lin Lin ◽  
Xiufang Liang

The online English teaching system has certain requirements for the intelligent scoring system, and the most difficult stage of intelligent scoring in the English test is to score the English composition through the intelligent model. In order to improve the intelligence of English composition scoring, based on machine learning algorithms, this study combines intelligent image recognition technology to improve machine learning algorithms, and proposes an improved MSER-based character candidate region extraction algorithm and a convolutional neural network-based pseudo-character region filtering algorithm. In addition, in order to verify whether the algorithm model proposed in this paper meets the requirements of the group text, that is, to verify the feasibility of the algorithm, the performance of the model proposed in this study is analyzed through design experiments. Moreover, the basic conditions for composition scoring are input into the model as a constraint model. The research results show that the algorithm proposed in this paper has a certain practical effect, and it can be applied to the English assessment system and the online assessment system of the homework evaluation system algorithm system.


2019 ◽  
Vol 1 (2) ◽  
pp. 78-80
Author(s):  
Eric Holloway

Detecting some patterns is a simple task for humans, but nearly impossible for current machine learning algorithms.  Here, the "checkerboard" pattern is examined, where human prediction nears 100% and machine prediction drops significantly below 50%.


Diabetes ◽  
2020 ◽  
Vol 69 (Supplement 1) ◽  
pp. 1290-P
Author(s):  
GIUSEPPE D’ANNUNZIO ◽  
ROBERTO BIASSONI ◽  
MARGHERITA SQUILLARIO ◽  
ELISABETTA UGOLOTTI ◽  
ANNALISA BARLA ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document